On 07/14/2016 08:17 PM, johnzeng wrote: > i deployed a set of squid 3.5.2 and redirect 300Mbps http traffic . > > workers 4 > cpu_affinity_map process_numbers=1,2,3,4 cores=1,3,5,7 > > > i check cpu status via top command ,and i found one of > squid is 92% at a time, whether it will be correct ? > > > Tasks: 292 total, 5 running, 287 sleeping, 0 stopped, 0 zombie > Cpu(s): 5.8%us, 1.2%sy, 0.0%ni, 89.4%id, 2.1%wa, 0.0%hi, 1.5%si, 0.0%st > Mem: 65889452k total, 64218068k used, 1671384k free, 138480k buffers > Swap: 29116412k total, 0k used, 29116412k free, 53204068k cached > > PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND > 1982 cacheuser 20 0 33.9g 1.3g 408m R 92 2.0 228:49.60 squid > 1983 cacheuser 20 0 33.9g 1.3g 408m R 56 2.1 254:23.07 squid > 1981 cacheuser 20 0 33.9g 1.3g 406m S 36 2.0 159:40.23 squid > 1984 cacheuser 20 0 33.5g 897m 408m R 33 1.4 189:14.00 squid The uneven distribution of work among workers [on Linux] is discussed at [1]. AFAIK, there were no significant relevant changes in Linux and in Squid since that wiki page section was written, but YMMV. [1] http://wiki.squid-cache.org/Features/SmpScale#Will_similar_workers_receive_similar_amount_of_work.3F HTH, Alex. _______________________________________________ squid-users mailing list squid-users@xxxxxxxxxxxxxxxxxxxxx http://lists.squid-cache.org/listinfo/squid-users
